Content uploaded by Hossein vahid dastjerdi
Author content
All content in this area was uploaded by Hossein vahid dastjerdi on Jan 13, 2015
Content may be subject to copyright.
ISSN 2039 - 2117
Mediterranean Journal of Social Sciences
Vol.2, No.2, May 2011
MCSER – Mediterranean Center of Social and Educational Research
Rome, Italy, www.mcser.org
65
Quality of Iranian EFL Learners’ Argumentative Essays: Cohesive
Devices in Focus
Hossein Vahid Dastjerdi
Samira Hayati Samian
University of Isfahan
English Department, Iran
h_vahid@yahoo.com; Sr.hayati.2006@gmail.com
Abstract
The ability to compose a piece of argumentative text is important for EFL and ESL learners. Despite its importance, there
is a gap in the literature about how Iranian students write essays in this genre that this study intends to fill. Building upon Halliday and
Hasan's (1976) cohesion theory, this study intended to investigate Iranian graduate non-English majors' use of cohesive devices in
argumentative essays, and also the relationship between the number of cohesive devices and writing quality. An analysis of forty
argumentative essays written by forty Iranian graduate non-English majors showed that the students were familiar with various cohesive
devices and used them in their writings. Among the cohesive devices used lexical devices had the largest percentage of the total number of
cohesive devices, followed by reference devices and conjunction devices. Furthermore, it was found that there was no significant relationship
between the number of cohesive devices used and quality of writing. The findings of the study have some important implications for EFL
writing teachers and learners.
Keywords:
Argumentative essays-
cohesive features- Iranian EFL learners
1. Introduction
Writing is one of the most authentic and interactive ways of transferring thoughts and ideas to
others. Halliday (1989) refers to writing as a negotiative and explonatory act, requiring great
judgement. The ability to express one’s ideas in writing in a second or foreign language
coherently and accurately is a major achievement that even many native speakers of English
never truly master it (Celce- Murcia, 2001). Learning to write efficiently a text is a long process
that requires much practice and sometimes explicit and formal instruction. For students who
have not yet acquired all the skills needed to translate their ideas into a coherent text, writing is
difficult and effortful. In recent years, researchers have given considerable attention to how EFL
and ESL learners actually write and what problems they usually encounter in their writing.
Several studies have indicated the problems that L2 writers have while writing (Chen, 2007;
Crewe, 1990; Kanno, 1989; Wu, 2006).
Learners’ writing must show some form of cohesion and coherence in their presentation
of ideas. At the discourse level, analysis of cohesion provides a useful measure of the
effectiveness and quality of written text. Since the publication of Cohesion in English by Halliday
and Hasan (1976), many researches have been made in the field of cohesion and coherence in
the English texts. Halliday and Hasan describe cohesion as one of the linguistic system's major
resources for text construction (p. vii). In fact, cohesion refers to the presence or absence of
explicit cues in the text that allow the reader to find relations of meaning within it. It is part of
the system of language which has the potentials for meaning enhancement in texts. In Wikborg’s
(1990) study, it was found that Swedish students often showed cohesion problems in their
writing ranging from missing or misleading sentence connection to malfunctioning cohesive
devices to too great a distance between the cohesive items in a cohesive chain.
In addition to having knowledge about the internal features of written texts, familiarity
with different genres can affect writing quality of the learners to a large extent. The notion of
ISSN 2039 - 2117
Mediterranean Journal of Social Sciences
Vol.2, No.2, May 2011
MCSER – Mediterranean Center of Social and Educational Research
Rome, Italy, www.mcser.org
66
genre is defined as “abstract, socially recognized ways of using language” (Hyland, 2003, p. 21)
which are purposeful communicative activities employed by members of a particular discourse
community (Swales, 1990). Argumentative writing is a fundamental writing style which is
required in higher education to compose various writing tasks. The goal of argumentative writing
is to convince an audience, and it is done in a situation where there exists a conflict between the
beliefs and attitudes of the writer/speaker and the reader/audience (Hinkel, 2002).
The purpose of the present study is to investigate cohesive devices used in argumentative
essays composed by Iranian graduate non-English majors, and the relationship between the use
of cohesive devices and quality of their essays.
2. Theoretical Background
Writing plays an important role in our personal and professional lives. Writing, by definition, is
an act of communication, a purposeful means of addressing an audience. However, writing is
currently viewed in academic circles as more than just a tool for communication. Therefore, the
ability to convey meaning proficiently in written texts is a critical skill for academic and
professional success. Indeed, college freshmen’ writing skills are among the best predictors of
academic success (Geiser & Studley, 2001), and even outside of academia, writing skills continue
to be important and are an important attribute of professional competence (Light 2001).
However, many students, particularly those attempting to write in their second language, rate
writing activities among the least enjoyable or beneficial for learning English (Barkhuizen, 1998;
Spratt, 2001). As such, developing a better understanding of characteristics of good writing is an
important objective, both for theoretical and applied reasons.
Cohesion and coherence, two important textual elements (Halliday and Hasan, 1976;
Halliday, 2000), have long been recognized as important features of ‘‘good’’ writing. In Halliday
and Hasan’s definition, coherence refers to the elements internal to the text, consisting of
cohesion and register: “A text is a passage of discourse which is coherent in these two regards: it
is coherent with respect to the context of situation, and therefore consistent in register; and it is
coherent with respect to itself, and therefore cohesive” (p. 23). Cohesion refers to the relations
of meaning that exists within a text, in other words, cohesion can be defined as linguistic devices
that are used to link one part of a text to another. Halliday and Hasan (1976: 04) mention that
cohesion occurs where the interpretation of some elements in the discourse is dependent on that
of another. Halliday and Hasan (1976: 04) note that concept of tie makes it possible to analyze a
text in terms of its cohesive properties and it gives a systematic account of its patterns and
texture. Cohesive ties can manifest in form of reference(i.e., the indication of information from
elsewhere such as personals, demonstratives, and comparatives), substitution(i.e., the
replacement of one component by another), ellipsis(i.e., the omission of a component),
conjunction (i.e., the indication of specific meaning which presupposes present items in the
discourse, such as additive, adversative, casual, and temporal)and lexical cohesion(i.e., the
repetition of the same or relative lexical items).
A number of researchers have investigated the relationship between the use of cohesive
devices and the overall quality of writing produced. However, the findings of these studies have
been somewhat inconsistent or contradictory. For example, some studies have contended that
there is a positive correlation between the number of cohesive devices and good writing (Cox
and Tinzmann 1987; Ferris, 1994; Field & Oi, 1992; Hasan 1984; Jin, 2001; Liu & Braine, 2005;
Pappas 1985). In contrast, other studies have not shown a significant relationship between the
number of cohesive features used and the quality of writing (Castro, 2004; Jafarpur, 1991;
Johnson, 1992; Neuner, 1987; Zhang, 2000). Some researchers also maintain that lexical devices
formed the largest percentage of the total number of cohesive devices, followed by references
and conjunctions in students' writings (Johns, 1980; Liu & Braine, 2005; Zhang, 2000).
ISSN 2039 - 2117
Mediterranean Journal of Social Sciences
Vol.2, No.2, May 2011
MCSER – Mediterranean Center of Social and Educational Research
Rome, Italy, www.mcser.org
67
In his study, Olateju (2006) investigated the extent to which ESL learners have been able to
achieve cohesion in their written texts by examining the cohesive devices used by the
students during their continuous writing sessions at school. The data used were drawn from
seventy final year students in a secondary school. The elicitation technique was an essay
writing exercise in which the students were given two essay questions which would enable
them to demonstrate their knowledge of cohesive devices in English. Although the students’
work showed evidence of the use of some of the cohesive devices identified by Halliday and
Hasan (1976 and 1985), some of the few used were wrong which made it difficult for
understand the texts. An analysis of the data revealed that the students lacked competence in
their use of cohesive devices despite the fact that they had been exposed to intensive teaching
of English for six years in the secondary school.
Writing in some genres is believed to be more difficult than writing in others.
The task of
constructing a successful piece of argumentative writing is complex and demanding conceptually
and structurally in comparison to composing a piece of narrative.
The writing of formal
argument places heavy cognitive demands on the writer. The organization of argument is more
difficult than the chronological order of narratives. Argumentative compositions by foreign
students often deviate characteristically from expected forms.
Producing the content of writing is also challenging. Argumentative texts are also considered
more difficult to write than narratives because they involve logical and coherent reasoning, which
are acquired late in cognitive development (Siegler, 1996). Studies in various countries have
reported poorer performance in argumentative writing than in other genres. In the United
Kingdom, major assessments of the writing of 11- and 15-year-olds conducted by the National
Foundation for Educational Research, found better performance on narrative writing than on
persuasive writing for both age groups, though the difference was not great (Gorman et al.,
1988).
However, the ability to compose a piece of argumentative text is considered important
for "academic success and for general life purposes" (Crowhurst, 1990, p. 349) and students need
to write dozens of lengthy papers before finishing their college careers.
Connor (1990) identifies four dimensions of argumentative texts that are unique to this
genre. Superstructure refers to the "organizational plan of any text and ... the linear progression of
the text" (p. 74). The second feature is the quality of logical reasoning which is assessed by
analyzing the interrelationships of writers' assertions and the associated support or data provided
to substantiate those claims. The third feature of good argumentative writing is identified as
persuasive appeal, including affective appeal and establishment of writer credibility. Finally, she
notes that audience awareness is an important characteristic of successful argumentative writing.
The writer must observe an awareness of the reader's perspective by "dealing implicitly or
explicitly with possible counterarguments" (p. 76).
Using Halliday and Hasan's (1976) analysis of elements in text cohesion, Crowhurst
(1981) examined differences in the argumentative prose written by 105 students in sixth, tenth,
and twelfth grade. The scale used to examine the students' writing contained 15 types of cohesive
ties. Crowhurst found a significant difference among grades for the frequency of four kinds of
cohesion: same lexical item, other lexical items, long-distance ties, and long-distance ties in the
last three T-units. These differences reveal that the older students were more likely to
foreshadow and summarize their arguments, thus producing long-distance ties and the repetition
of the same lexical items in summaries. In addition, the older students, who have larger
vocabularies, more frequently used a variety of terms to refer to the same concept. In this study
Crowhurst found no significant difference among grade levels in the number of students using
the various types of cohesion. Crowhurst notes that the finding of differences in only 4 of the 15
types of cohesion does not reveal very much about the distinctions in argumentative writing
ability among the three age groups.
ISSN 2039 - 2117
Mediterranean Journal of Social Sciences
Vol.2, No.2, May 2011
MCSER – Mediterranean Center of Social and Educational Research
Rome, Italy, www.mcser.org
68
In her study, Castro (2004) compared the degree of cohesion and coherence in the essays written
by thirty Filipino college freshmen. Though the results of the cohesion analysis indicated no
significant difference in the number and types of grammatical or lexical cohesive devices in the
low, mid and highly rated essays, the results can be useful for writing teachers who can teach the
students the appropriate cohesive markers and emphasize their importance in writing.
Using Stephen Toulmin’s (2003) model of argument, Chen and Cheng (2009) examined
the use of English argumentation features in Taiwanese and US college freshmens’ writings. The
findings indicated that Taiwanese student arguments were less extended and complex, and
displayed a limited range and quantity of argumentative structure in comparison to American
arguments. Yet, both Taiwanese and American students were weak at handling oppositional
structures, an essential trait differentiating Chinese and English rhetoric. Equally important,
Taiwanese students, when composing Chinese texts were able to construct certain argument
features in a way similar to American students. This illustrated that culture may not necessarily
account fully for the argument features manifested in Taiwanese writing of English. Other
factors, such as L2 language proficiency and developmental factors also played a mediating role
in the use of argument structures.
McNamara et al (2010) used linguistic indices of cohesion and language sophistication
provided by the computational tool Coh-Metrix (Graesser, McNamara, Louwerse, & Cai, 2004)
to analyze a corpus of 120 argumentative essays written by college undergraduate and scored by
expert raters using a holistic rubric. The essays were scored on a 1-6 scaled SAT rubric and then
categorized into two groups: essays judged as low versus high quality. The results indicated that
there were no differences between these two groups according to indices of cohesion (e.g., word
overlap, causality, connectives). By contrast, indices related to language sophistication (lexical
diversity, word frequency, and syntactic complexity) showed significant differences between the
groups.
Wang and Cho (2010) examined two major academic genres of writing: argumentative
and technical writing. Three hundred eighty-four undergraduate student-produced texts were
analyzed through a computational tool called Coh-Metrix. The results showed that students used
genre-dependent cohesive devices in a limited way to write research papers. Students' writings
were examined in seven dimensions of textual cohesion. For instance, it was found that students'
argumentative writing texts tend to have complex syntactic structures (due to overuse of pre-
modifications) that affect the cohesion of texts. Furthermore, students employed impersonal
constructions (passive voice) in their technical writing; however, over-indulgence in passive voice
may also cause ambiguous meanings. As to the causal cohesion dimension, the results suggested
that college students write argumentative writing with more causal cohesion than for their
technical writing. On the dimension of connectives, it was found that college students'
argumentative texts include significantly more connectives in their argumentative writing than in
their technical writing, excepting the use of positive temporal connectives. In addition, it was
found that students' technical writing showed significantly higher co-referential cohesion than
students' argumentative writing for all measures. At the dimension of density of part of speech,
students used personal pronouns are used less in their technical writing than in their
argumentative writing. Regarding syntax complexity, students used significantly more modifiers
in their technical writing than in their argumentative writing. Finally, students-produced
argumentative texts are more difficult to comprehend than their technical writing texts.
Many researchers have explored the relationship between the use of cohesive devices and
the quality of the writing produced. However, the results of various studies have been at best
inconsistent. In order to shed some light on this area of debate, the present study investigated
Iranian non-English major graduates' use of cohesive devices in argumentative writing and the
relationship between the number of cohesive devices and quality of their writing. Argumentative
writing has been chosen as the focus of the present study since it is important both for academic
success and for general life purposes. University students must be able to write argumentative
ISSN 2039 - 2117
Mediterranean Journal of Social Sciences
Vol.2, No.2, May 2011
MCSER – Mediterranean Center of Social and Educational Research
Rome, Italy, www.mcser.org
69
papers in order to join the conversation in the academic community. Despite its importance,
there is no accurate picture of how Iranian students compose English argumentative writing and
how effective their English argumentation is.
Having reviewed previous studies on argumentative writing and cohesive devices, this
study was conducted in an EFL context, Iran, to address the following questions:
1.
What types of cohesive devices are used by Iranian graduate non-English majors in their
argumentative writing?
2.
How frequently do Iranian graduate non-English majors use cohesive devices?
3.
Is there a significant relationship between the number of cohesive devices used and the
quality of writing produced?
3. Method
3.1. Participants
The participants in this study were 40 Iranian EFL graduate non-English major who enrolled in
an English Writing course at a private language institute in Isfahan. None of them had any
formal instruction about writing before starting this course. The participants came from different
departments ranging from Law to Medical sciences to Statistics. All of them had passed a
placement test to be at the expected level of proficiency before starting the course. They were
taught basic writing skills and different genres and styles. At the end of the course, they were
expected to be able to write different types of essays including narrative and argumentative ones
among others. The number of male and female participants was equal, 20 female and 20 male.
Their age range was 22 to 30.
3.2. Data Collection Procedure
At the first session, objectives and requirements of the course were clearly specified for the
students. Therefore, the students found out what they should do during the course and what is
expected of them at the end of the course. During the term, the students were taught principles
of rhetoric and organization, provided with a text of a specific genre for classroom discussion,
analysis, and interpretation, were required a writing assignment of that genre, and finally their
writings were read and commented by their teacher. In addition, the concepts of cohesion and
coherence were explicitly emphasized in each sample throughout the course. Therefore, the
students had a clear idea about how they should compose their writing assignments. In the final
session, the researcher asked them to write an argumentative essay on the topic, "Drinking a lot
of water can help you to become healthier" within 25 minutes. The participants were required to
state their viewpoints about this topic and defend it.
3.3 Data Analysis Procedures and Results
Following Liu & Braine’s study (2005) and Zhang’s (2000)’s study, the data in this study were
analyzed through two procedures: identifying and counting the number of cohesive devices and
evaluating the quality of written texts. For the first phase of data analysis, Halliday and Hasan's
(1976) cohesion taxonomy was used to examine the kinds of cohesive ties used in argumentative
texts. Then, frequency, mean and standard deviation of the cohesive devices in each category
were computed using SPSS statistical software package. However, two categories of cohesive
devices, substitution and ellipsis, were not analyzed because ‘‘they are more characteristically
found in dialogues’’ (Halliday, 2000, p. 337) and “they are seldom used in formal writing” (Liu &
Braine, 2005, p. 647).
ISSN 2039 - 2117
Mediterranean Journal of Social Sciences
Vol.2, No.2, May 2011
MCSER – Mediterranean Center of Social and Educational Research
Rome, Italy, www.mcser.org
70
For the second and the qualitative phase of the study, the argumentative texts were
independently rated by two raters, i.e., the researcher and another English teacher who had at
least 4 years of experience teaching composition classes. A holistic rating scale (Educational
Testing Service, 2004) ranging from zero to five points was used. Then, the inter rater reliability
was calculated to show the consistency of scoring. The inter-rater reliability was .813, which
indicated the overall writing scores were consistent and reliable. Then, the relationship between
the frequency of the use of cohesive devices and the quality of writing was examined through the
use of Pearson Correlation.
Forty essays were evaluated by the two raters for the present study. The scores given to
each composition by the raters were averaged and the mean was determined as the final score for
that piece, and the averaged scores were correlated with the number of cohesive devices used by
students to reveal the potential relationship between the numbers of cohesive devices and
writing quality. The results are presented in Table-1below:
Table 1
.
Mean, Standard Deviation, Range, and Other Statistics Related to the Argumentative text Scores
Mean
Standard deviation
(SD)
Standard error (SE) Minimum
Maximum Range Median
3.26 0.6 0.1 2.50 4.50 2.00 3.12
As shown in Table-1, the mean score of the 40 compositions is 3.26 (out of a maximum score of
5) and the standard deviation is 0.6. Thus, the compositions scored four points or above were
considered the best, while those scored three or below were regarded the weakest. The range of
distribution of score was narrow. This may indicate that the students in this study are far apart in
terms of writing ability, and also it can be concluded that the participants’ average writing
proficiency was at high-intermediate level.
3.3.1. Cohesive Devices Used in Argumentative Text
The type and number of cohesive devices used in each argumentative text were analyzed using
Halliday and Hasan’s (1976) cohesive framework as the basis for data analysis. Table 2 illustrates
the frequency, mean per essay, standard deviation, and percentages of the different subcategories
of cohesive devices identified in the texts. This shows that the students in the present study used
a variety of cohesive devices, and they employed some types of devices more frequently than
others. From the data it is evident that the participants used lexical devices (52.2%) more
frequently than reference devices (27.6%) followed by conjunction devices (20.2%).
Table 2. Cohesive Devices Used in Essays
Type of cohesive
devices
Reference devices
Conjunction
devices
Lexical devices Total number of
cohesive devices
Frequency 1154 844 2186 4184
Mean per composition 28.85 21.2 54.65 104.7
Standard deviation 5.95 3.84 15.97 10.91
Range
53 29 68 68
Percentage based on
total
27.6% 20.2% 52.2% 100%
ISSN 2039 - 2117
Mediterranean Journal of Social Sciences
Vol.2, No.2, May 2011
MCSER – Mediterranean Center of Social and Educational Research
Rome, Italy, www.mcser.org
71
These findings concerning the frequency order of different types of cohesive devices are
compatible with that of previous research studies (Liu & Brain, 2005; Neuner, 1987; Yvette and
Yip, 1992; Zhang, 2000). Following Liu & Braine’s study (2005), the definite article ’the’ was also
calculated and integrated into reference devices. Therefore, higher percentage of reference
devices used may be due to the inclusion of the definite article. To pinpoint the variety of
different subcategories of cohesive devices, a more detailed analysis of them is presented below.
3.3.2. Reference Devices Used in Essays
The data displayed in table 3 indicate that among the three sub-categories of reference devices,
pronominal devices (51.3%) occupied the largest percentage of use, followed by the definite
article (26.7%), the comparatives (12%), and demonstratives (10%) which had the least
percentage of use.
Table 3. Reference Devices Used in Essay
Total number
of reference
devices
Comparative
devices
The definite
article
Demonstratives
Pronominal
devices
Reference
devices
1154 137 309 115 593 Frequency
28.83 3.42 7.72 2.87 14.82 Mean per essay
5.95 1.86 3.38 1.81 5.69 Standard
deviation
53 9 17 11 26 Range
100% 12% 26.7% 10% 51.3% Percentage
Mor e- muc h-
le ss- the
mo s t
_ There- that-
this
You- we- I
your- our
Most frequently
used cohesive
Items
The findings correspond to Liu & Brain (2005), but differ from Neuner (1987) in which the
number of demonstratives was slightly higher than the definite article. In terms of the most
frequently used pronouns, ‘you’ was used the most followed by ‘I’ and ‘we’. It might reflect the
fact that most of the students were more comfortable at using first and second persons to make
their writings more subjective and personal. This observation shows that the students should be
taught to use third person pronouns in order to make their argumentative writings more
objective and authoritative. Among demonstratives, students used ‘this’ and ‘that’ more than
‘these’ and ‘those. It shows that the students preferred to use singular demonstratives more than
plural ones. The following are some of the extracts taken from the students’ essays:
Example 1
What if someone told you, you could lose weight with no effort on your part? What if
there was a secret to losing weight that didn’t involve our exercise or decrease calorie.
Example 2
The secret to much fast weight loss without exercise or diet changes is to drink more
water. I think this can also make you more healthier.
3.3.3. Conjunction Devices Used in Essays
Table 4 demonstrates the frequency, percentage and standard deviation of the four
ISSN 2039 - 2117
Mediterranean Journal of Social Sciences
Vol.2, No.2, May 2011
MCSER – Mediterranean Center of Social and Educational Research
Rome, Italy, www.mcser.org
72
subcategories of conjunction devices. Among these subcategories, additive devices (51.2%) had
the largest percentage of use, followed by the Causal devices (19.3%), adversative devices
(15.5%), and temporal devices (14%). These findings are somewhat different from those of Liu
& Brain (2005) in which the number of temporal devices was higher than the adversative
devices.
In terms of the most frequently used conjunction devices, it is interesting to find that in
each category, the students in this study strongly preferred using simple words to longer phrases
to connect different parts of their writing together. The cohesive items with the highest
frequency among additive devices were ‘and’, ‘or’, and ‘also’. Among adversative devices, the
students employed item ‘but’ and ‘however’ the most frequently, whereas they rarely used items
like ‘on the contrary’ and ‘on the other hand’. In terms of causal devices, the items ‘because’,
‘because of’, and ‘for’ had the highest percentage. As to temporal, the students employed ‘first’,
‘second’, ‘finally’, and ‘at the end of’ more than others to show the order of their reasoning. It
might manifest the fact that most of them lacked familiarity with or had difficulty with using
other conjunction devices. This observation could inform writing teachers to emphasize complex
conjunction devices more in their teaching.
Table 4. Conjunction devices used
Total number
of conjunctive
devices
Temporal
devices
Causal devices
Adversative
devices
Additive
devices
Conjunction
devices
844 118 163 131 432 Frequency
21.09 2.95 4.07 3.27 10.8 Mean per essay
3.84 1.06 2.09 1.43 3.16 Standard
deviation
29 4 9 6 13 Range
100% 14% 19.3% 15.5% 51.2% Percentage
First, second,
finally, at the
end
Because,
because of, for
But, however
And, or, also
Most frequently
used cohesive
Items
3Example
it , necessary component of your bodyand water is such an important because, Finally
makes sense that it can play a vital role in your health and body conditions.
4Example
... it can clean thembecause, helpful for our kidneysandwater is very useful , First
it helps to lose our Also.more weight loss and drinking water leads to less eating, Second
. we don’t feel hungrybecause, appetite
3.3.4 Lexical Devices Used in Essays
Among the five sub-categories of lexical devices, repetition (76%) accounted for the largest
percentage of use, followed by synonym (8.7%), collocation (7.9%), antonym (4.8%), and
sperordinate (2.6%), as shown in Table 5. The results are in line with Liu & Brain (2005), Neuner
(1987), and Zhang (2000) studies, and indicate that the students had a tendency to use the same
vocabulary item to emphasize their ideas and support their argument. It can be due to the fact
that the students have a limited knowledge of vocabulary; hence they should repeat them in their
ISSN 2039 - 2117
Mediterranean Journal of Social Sciences
Vol.2, No.2, May 2011
MCSER – Mediterranean Center of Social and Educational Research
Rome, Italy, www.mcser.org
73
writings. As can be seen from Table 5, the most frequently repeated vocabulary items were
‘water’, ‘body’, ‘drink’, and ‘weight’ which were directly related to the topic of the essay. The
followings are some extracts from students’ essays:
Table 5. Lexical devices use
Lexical devices
Repetition Antonym Synonym Superordinate
Collocation
Total
number of
lexical
devices
Frequency 1663 104 189 58 172 2186
Mean per essay
41.57 2.6 4.72 1.45 4.3 54.64
Standard
deviation
9.11 0.90 1.86 0.63 1.93 15.97
Range 49 4 9 3 11 68
Percentage 76% 4.8% 8.7% 2.6% 7.9% 100%
Most
frequently used
cohesive
Items
Water-
calorie-
body-
human-
drink
Easy-
difficult
Useful-
harmful
Advantage-
disadvantage
Get-obtain
Exercise-
activity
Liquid-water
Exercising-
running
Put on
weight-lose
weight
Do exercise
3.3.5. The Relationship Between the Number of Cohesive Devices and Writing Quality
In addition to the general description of the frequency of use of cohesive devices, another major
purpose of this study was to investigate relationship between the number of cohesive devices
used and the quality of writing produced. To answer research question 3 regarding this point, the
numerical essay scores and the number of each cohesive category (i.e., total devices per
composition) were correlated by Pearson Product Moment Correlation Coefficient.
Table 6. Correlation between Essays Scores and Cohesive Devices
Essay scores
Reference
devices
Conjunction
devices
Lexical
devices
Total
number of
cohesive
devices
Essay scores
1
Reference devices
-0.306 1
Conjunction devices
-0.026 0.710(**)
1
Lexical devices
-0.201 0.803(**) 0.857(**) 1
Total number of
cohesive devices
-0.306 0.887(**) 0.710(**) 0.686(**)
1
*p<0.05
**p<0.01
ISSN 2039 - 2117
Mediterranean Journal of Social Sciences
Vol.2, No.2, May 2011
MCSER – Mediterranean Center of Social and Educational Research
Rome, Italy, www.mcser.org
74
The correlation matrix in Table 6 indicates non-significant and even negative correlations
between (1) the essay scores and the number of reference devices (r = - 0.306), (2) the essay
scores and the number of conjunction devices (r = -0.026), (3) the essay scores and the number
of lexical devices (r =-0.201), and (4) the essay scores and the total number of cohesive devices (r
= -0.306). It is evident that the number of cohesive devices and writing quality were shown not
to be significantly related. The greater use of cohesive devices in writing did not indicate better
writing quality.
One possible explanation may be that the participants in this study overused cohesive
devices and in some cases even misused them in their essays. It seems logical to conclude that
their problems with using the cohesive devices had negative effect on their writing quality.
Analyzing their essays revealed that most of them didn’t know how to use those cohesive devices
properly. Some of the participants’ problems were related to the use of reference devices. For
instance, regarding the use of definite article ‘the’, it seemed to be some interference from the
Persian language. The participants tended to confuse the use of definite and indefinite articles or
insert unnecessary ones. In addition, in some cases, they employed double-comparative
structures (e.g., the more healthier or the least harmfulest or the more better…). Apart from the difficulties
with reference devices, the students tended to
overuse additive devices (e.g., and, or...) and causal
devices (e.g., because, because of...) in their essays. Participants sometimes used such causal devices
without any clear cause- effect relationships among parts of sentences. A more salient problem
was initial positioning of conjunction devices even when it is allowed to place them in non-initial
position. That is, most of conjunctive words or phrases appeared at the beginning of sentences.
Similar finding was found in Zhang’s study (2000). Lexical devices were the most problematic
area for the participants in this study, even though they accounted for the highest percentage of
cohesive devices identified in argumentative essays. The students used a limited number of
lexical items and most of the lexical devices were just repetitions of the same items. Other types
of lexical devices were rarely used in the writing essays. Furthermore, analysis of data revealed
that the participants had difficulty choosing right words and right collocations (e.g., miss some
weight). There may be two sources for the students’ problems in this regard. One is that limited
exposure to authentic materials and related readings. The other one is the interference from
mother to second language.
4. Discussion and Conclusions
The findings of this study, in general, are consistent with those of previous research studies. In
this study, Halliday and Hasan’s (1976) cohesive framework was used to analyze students’ use of
cohesive devices. To summarize, the results of this study suggest that the participants, Iranian
graduate non-English majors, had knowledge of cohesive devices and were capable of employing
a variety of them in their argumentative writings. However, some of the cohesive devices
employed were wrongly used which made it difficult to comprehend the text. Among the three
cohesive devices examined, lexical devices (52.2%) formed the highest percentage of the total
number of cohesive devices used in the argumentative essays, followed by reference devices
(27.6%) and conjunctions (20.2%).
A more detailed analysis of the cohesive devices used in the argumentative essays showed
that, in reference devices category, pronominal devices (51.3%) were the most frequently used
while demonstratives (10%) the least frequently used. This finding corresponds to that of Liu &
Brain’s study (2005), while the percentage of use is a little different. Regarding the use of
conjunctions, the qualitative analysis of the essays indicated that the participants of this study
preferred using simple conjunctions like ‘and’, ‘but’, ‘because’ more frequently than others like
‘nonetheless’, ‘on the other hand’, ‘furthermore’. It may be due to the fact that the students learn
the simple ones in early stages of second language learning, hence feel more comfortable using
them. It was found that Iranian graduate non-English majors in this study were in general weak
ISSN 2039 - 2117
Mediterranean Journal of Social Sciences
Vol.2, No.2, May 2011
MCSER – Mediterranean Center of Social and Educational Research
Rome, Italy, www.mcser.org
75
in lexical cohesion, though it constituted the highest percentage of total cohesive devices used in
the essays. Extensive use of lexical devices was reasonable because this genre requires forming
arguments and elaborating ideas to support them which necessitate the effective use of various
lexical devices. The students ‘vocabulary repertoire was limited. It can be understood by the fact
that a great percentage of the lexical devices was merely repetition of the same lexical items
(76%). In addition, some of the lexical items used, especially the collocations, were misused. This
finding is in line with that of Liu & Brain’s study (2005).To understand any relationship between
the frequency of cohesive devices used and the quality of writing, correlation was computed
between the numerical essay scores and total number of cohesive devices in essays. The
correlation matrix indicated that there was no statistically significant relationship between the
essay scores and the number of cohesive devices used in the same essay. This finding is in line
with those of the previous research studies (Castro, 2004; Jafarpur, 1991; Johnson, 1992; Karasi,
1994; Neuner, 1987; Zhang, 2000). Therefore, the number of cohesive devices cannot be a
differentiating factor between ‘good essays’ and ‘poor ones’.
Generally speaking, lack of cohesion in writing is a problem that cannot be ignored. The
findings of this study suggest the following implications both for writing teachers and EFL
learners. The first one is for the students to improve their lexical knowledge. Acquiring lexical
knowledge is fundamental in learning the four skills of the second or foreign language. Inclusion
meaningful vocabulary exercises in writing course syllabuses may address this issue. In addition,
new vocabulary items should not be presented in isolation rather in context, since it may help the
learners to distinguish the differences between them. Well-organized lexical knowledge can help
the learners to compose more lexically cohesive essays.
Second, as most of the students in the sample were found to have difficulty employing the
cohesive devices accurately and properly, it seems necessary to teach cohesion and cohesive
devices explicitly and to provide them with ample examples in English classes. Since in most
cases, learners are familiar with different types of cohesive devices; however, they simply do not
know how to use them correctly.
Third, future research should be done to consider the growth of cohesion knowledge that results
from a variety of instructional sources and its time in learners. This is especially important to
consider because the time of learning about cohesion needs to be identified in order to develop
instructional programs to facilitate such learning.
A final word is that although this study did not analyze a large number of essay samples,
it may be considered a helpful contribution, particularly in our country, Iran, where few studies
have examined cohesion in Iranian EFL learners’ writings, and especially the relationship
between use of cohesive devices and (argumentative) writing quality.
References
Barkhuizen, G.P. (1998). Discovering learners' perceptions of ESL classroom teaching/learning activities in a South
African context. TESOL Quarterly, 32(1), 85-108.
Castro, C. (2004). Cohesion and the social construction of meaning in the essays of Filipino college students writing
in L2 English. Asia Pacific Education Review, 5(2), 215-225.
Celce-Murcia, M. (Ed.). (2001). Teaching English as a second or foreign language (3rd edition.). Boston: Heinle & Heinle.
Cheng, F., Chen, Y. (2009). Taiwanese argumentation skills: Contrastive rhetoric perspective. Taiwan International
ESP Journal 1(1), 23-50.
Chen, Y. L., & You, Y. L. (2007). Less experienced EFL writers’ knowledge and self-awareness of coherence in
English writing. Selected Papers from the Sixteenth International Symposium and Book Fair on English Teaching, 335-
346. Taipei: Crane.
Connor, U. (1990). "Linguistic/Rhetorical measures for International persuasive student writing”. Research in the
Teaching of English, 24, 67-87.
Cox, B.E., & Tinzmann, M. (1987) Elementary children's knowledge of exposition. Paper presented at the annual
meeting of the National Reading Conference, St. Petersburg, FL.
Crewe, W. J. (1990). The illogic of logical connectives. ELT Journal, 44(4), 316-325.
Crowhurst, M. (1981). Cohesion in argumentative prose written by sixth-, tenth- and twelfth- graders. Paper
ISSN 2039 - 2117
Mediterranean Journal of Social Sciences
Vol.2, No.2, May 2011
MCSER – Mediterranean Center of Social and Educational Research
Rome, Italy, www.mcser.org
76
presented at the annual meeting of the American Educational Research Association, Los Angeles,
California. (ERIC Document Re-production Service No. ED 202 023)
Crowhurst, M. (1990). Teaching and learning the writing of persuasive/argumentative discourse. Canadian Journal of
Education, 15 (4), 348-359.
Educational Testing Service (2004). iBT/Next Generation TOEFL Test: Integrated Writing Rubrics Retrieved June 25,
2008, from http://www.ets.org/Media/Tests/TOEFL/pdf/Writing_Rubrics.pdf
Ferris, D. R. (1994). Lexical and syntactic features of ESL writing by students at different levels of L2 proficiency.
TESOL Quarterly, 28(2), 414-420.
Field, Y., & Oi, Y. L. M. (1992). A comparison of internal conjunctive cohesion in English essay writing of
Cantonese speakers and native speakers of English. RELC Journal, 23, 15-28.
Geiser, S. & Studley, R. (2001). UC and SAT: Predictive validity and differential impact of the SAT I and SAT II at the
University of California. Oakland, CA: University of California.
Gorman, T.P., White, J., Brooks, C., MacLure, M., & Kispal, A. (1988). A review of language monitoring 1979-83.
London: HMSO, Assessment of Performance Unit.
Graesser, A.C., McNamara, D.S., Louwerse, M.M., & Cai, Z. (2004). Coh-Metrix: Analysis of text on cohesion and
language. Behavior Research Methods, Instruments, and Computers, 36, 193–202.
Halliday, M.A.K., Hasan, R., (1976). Cohesion in English. Longman, London.
Halliday, M., and R. Hasan (1985). Language, context, and text: Aspects of language in a social-semiotic perspective. Victoria:
Deakin University Press.
Halliday, M.A.K.,( 2000). Introduction to Functional Grammar, second ed. Foreign Language Teaching and Research Press,
Beijing.
Hassan, R. (1984). Coherence and cohesive harmony. In J. Flood (Ed.), Understanding reading comprehension (pp. 181-
219). Newark, DE: InternationaRl eadingA ssociation.
Hinkel, E. (2002). Second language writers’ text. Mahwah, NJ: Erlbaum.
Hyland, K. (2003). Second language writing. Cambridge: Cambridge University Press.
Jafarpur, A., (1991). Cohesiveness as a basis for evaluating compositions. System 19, 459–465.
Jin, W. (2001). A quantitative study of cohesion in Chinese graduate students’ writing: Variations across genres and
proficiency levels. ERIC Document Reproduction Service, No. ED 452 726.
Johns, A.M., (1980). Cohesion in written business discourse: some contrasts. The ESP Journal 1, 36– 44.
Johnson, D.P., (1992). Cohesion and coherence in compositions in Malay and English. RELC Journal 23, 1–17.
Kanno, Y. (1989). The use of connectives in English academic papers written by Japanese students. Psycholinguistics,
2, 41-54.
Karasi, M. (1994). Cohesive features in expository essays of Secondary Four (Express) and Secondary Five (Normal)
students in Singapore. M.A. dissertation. Nanyang Technological University; National Institute of
Education.
Light, R. (2001). Making the most of college. Cambridge, MA: Harvard University Press.
Liu, M., & Braine G. (2005). Cohesive features in argumentative writing produced by Chinese undergraduates.
System, 33, 623-636.
McNamara, D.S., Louwerse, M.M., McCarthy, P.M., & Graesser, A.C. (2010). Coh-Metrix: Capturing linguistic
features of cohesion. Discourse Processes, 47, 292-330.
Olateju, M. (2006). Cohesion in ESL Classroom Written Texts. Nordic Journal of African Studies 15(3), 314–331.
Pappas, C. C. (1985). The cohesive harmony and cohesive density of children's oral and written stories. InJ. D.
Benson & W. S. Greaves (Eds.), Systemic perspectives on discourse: Vol. 2. Selected applied from the
ninth International Systemic Workshop (pp. 169-186). Norwood, NJ: Ablex.
Neuner, J. L. (1987). Cohesive ties and chains in good and poor freshman essays. Research in the Teaching of English, 17,
215-229.
Siegler, R. S. (1996). Emerging minds: The process of change in children’s thinking. New York:Oxford University Press.
Spratt, M. (2001). The value of finding out what classroom activities students like. RELC Journal, 32, 80-103.
Swales, J. M. (1990). Genre analysis: English in academic and research settings. Cambridge: Cambridge University Press.
Toulmin, S. E. (2003). The uses of argument (2nd ed.). Cambridge, UK: Cambridge University Press.
Wang, X., Cho, K. (2010). Computational Linguistic Assessment of Genre Differences Focusing on Text Cohesive
Devices of Student Writing: Implications for Library Instruction. The Journal of Academic Librarianship, 36(6),
501- 510.
Wikborg, E., (1990). Types of coherence breaks in Swedish student writing: misleading paragraph division. In:
Connor, U., Johns, A.M. (Eds.), Coherence in Writing: Research and Pedagogical Perspectives. TESOL,
Alexandria, VA, pp. 131–149.
Wu, S. R. (2006). Connectives and topic-fronting devices in academic writing: Taiwanese EFL student writers vs.
international writers. 2006 International Conference and Workshop on TEFL and Applied Linguistics, 417-425.
Yvette, F., Yip, L., (1992). A comparison of Internet conjunctive cohesion in the English essay writing of Cantonese
speakers and native speakers of English. RELC Journal 23, 15–28.
Zhang, M., (2000). Cohesive features in exploratory writing of undergraduates in two Chinese universities. RELC
Journal 31, 61–93.
